Snakes and ladders
Oli Gould, design team manager, Owen Mumford examines the ‘snakes and ladders’ of medical device design and development.
There lies the famous expression “you don’t know what you don’t know”, and never has a truer sentence been spoken. US Secretary of Defense, Donald Rumsfeld, was correct in stating there will always be things we know about and can prepare for, but equally there will also be times when we face the “unknown unknowns” - the things we don’t know about. It is these things, mixed with the “known unknowns”, which make the strategy applied to design development such a challenging topic.
In medical device manufacturing, the design process can present itself as a chaotic system unless it is managed well. Customer requirements can encapsulate a range of material and design requirements and whilst the design team may be well equipped to handle this, there can be vast differences between the idea on paper and its end result in physical form and there can be many unexpected challenges along the way.
The ‘Snakes and Ladders’ philosophy
For design engineers, medical device design and development can offer many similarities to the classic game of ‘Snakes and Ladders’.
The ‘Ladders’ on the board represent all the big wins that could be uncovered during the development stages. This could be the unique selling points, the “if we do this we can increase performance by X%!” statements and the “if we combine the function of these parts into one, we can save X% in cost!”
In comparison, the ‘Snakes’ on the board represent all the big failures that could befall the concept design. They are the major delays, the “how did we miss this?”, the “who would have thought that this would happen?”
A stage of development activity which seeks to assess or prove the design concept can be likened to a move in the board game. In a poorly planned development strategy, the team may pass over a Ladder, or a Snake, without even knowing it was there. The team may approach the Finish only to find the top of a Ladder that they hadn’t landed on before, or the head of a Snake which takes them back to the start of their project.
It is the role of a good development strategy to employ the tools, processes and techniques to detect all the Snakes and the Ladders as early as possible when developing a device. Whatever the journey looks like, there is no such thing as a simple journey from A to B.
Detecting the Snakes
Whilst navigating the board, it’s finding the biggest Snakes or failures, those which could cause major redesign, which should be the priority. Smaller failures cannot be ignored either because, in combination, they may cause an effect on a similar scale. The ‘known unknowns’ are often the easiest to detect as the conditions and areas of focus are known at the outset. The ‘unknown unknowns’ are often more difficult to detect as the conditions of failure and failure modes are not known.
At Owen Mumford, our vast experience in design for manufacturing enables us to foresee any potential challenges or Snakes across the product journey, but we cannot determine them all. We ensure that all our devices, whether made up of known parts or new, are subject to rigorous evaluation.
The following are some of the key methods we employ at Owen Mumford to detect the Snakes:
Fail fast
In engineering, we encourage a fail-fast system whereby we apply a great deal of effort to flush out problems as early as possible. This means starting activities which are traditionally undertaken at later stages much earlier, such as manufacturability assessments, assembly equipment design and measurement system accuracy (MSA) analysis.
The ‘what if?’
The ‘known unknowns’ are easier to detect and understand than any ‘unknown unknowns’ and we can employ techniques to convert from one to the other. Since the only difference between these is an awareness of missing information, tools such as task analysis, function tree and fault tree analyses help identify the ‘unknown unknowns’ by walking step by step through the device operation.
Make and test
To detect the ‘unknown unknowns’, physical testing is one of the very best methods as it removes the opportunity for erroneous assumptions that may be present in other detection means, such as virtual and mathematical modelling. Physical testing is not infallible as assumptions may be made under the conditions of manufacture and the conditions of test. At Owen Mumford, we employ various stages of ‘make and test’, from feature level through to full device.
Make and trial
Whilst some analysis methods associated with product use can detect some potential failures, trials with actual users in situations as close to reality as possible are invaluable in detecting the ‘unknown unknowns.’
Model and simulate
Whilst models and simulations of a device or mechanism can often have some form of assumption or simplification included (and risk that some may be inappropriate), they also provide a fantastic opportunity to explore possible combinations of manufacturing and environmental conditions, in a fraction of the time and cost that it would take to do it physically.
Detecting the Ladders
It may be obvious that one first considers the best way to approach something before doing it. Once a route has been decided upon however, it can be all too easy to accelerate to the finish line but during the process of development, much of the design is in flux. Decisions are being made on how the design should proceed so that Snakes are avoided. An engineer must also take stock at regular intervals to make sure the combination of those decisions is the most effective in terms of product performance.
During the initial stages of development, identifying ways in which the end product performance will excel (i.e. the Ladders) is critical to success. It is the hallmark of many successful medical organisations.
There are numerous tools available to an engineer to create innovative and varied solutions to meet a set of requirements. There is complexity and subtlety in this ladder building exercise; it must balance flexibility with structure in its approach. Tools such as Brainstorming, TRIZ, Systematic Inventive Thinking and Six Thinking Hats are just some of the methods to assist at the early stages.
Later in development, when the concept starts to take shape, an engineer needs to be conscious of the decisions which have led to the challenge at hand and the information and assumptions used in those decisions. He or she must assess whether the decisions fundamentally and on balance still hold true because development challenges within the project represent opportunities for added complexity (which may be warranted) or continued simplicity. Simplicity is often correlated with timely delivery and good product performance.
Using tools such as a Decision Tree, it is possible to detect trends in complexity which can be traced back to an earlier decision. Taking a ‘time out’ to re-evaluate that decision and on occasion, taking the hard choice to switch direction is a crucial part of delivering a high-performance solution and can often save time compared to the alternative, more complex route.
Navigating the board game
These tools and techniques can enable you to move through the board but how you use these methods is also important.
An optional methodology is to pursue full detail in these methods right from the start. However, when a Snake or Ladder is detected, the design changes and the detection processes must be started again. This methodology is akin to meticulously stepping through the board one space at a time. When you arrive at the finish, you will have certainly uncovered every Snake and every Ladder but you will have spent an exorbitant amount of time and money to get there, impacting product timelines along the way.
At Owen Mumford, we apply a development strategy which advances selective aspects of the process forward, whilst others remain behind, to probe the spaces ahead of us on the board. By investing time in selectively probing ahead, we invest a little to save a lot.
The selection of factors to probe come mostly from engineering knowledge, comparable products, and from findings from previous probes. The factors selected are analysed so a level of detail commensurate with the severity of impact of a possible Snake (or Ladder) on the ability to deliver a solution, but balanced against the time one must invest to attain that level of detail.
As you approach the end of the game, or the end of device development, you will have adapted your design based on the Snakes and Ladders you have experienced along the way. These ‘realities’ can cause hurdles, but the approach undertaken here at Owen Mumford enables a robust conclusion to the development process, where you have identified and managed all the problems and considered all of the opportunities. Product development is a continuous journey of feedback and modification and it’s how you respond to the challenges that count.